The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> change the position of Rc&Cc for the miller effect
https://designers-guide.org/forum/YaBB.pl?num=1320888686

Message started by henrytqy on Nov 9^th, 2011, 5:31pm

Title: change the position of Rc&Cc for the miller effect
Post by henrytqy on Nov 9^th, 2011, 5:31pm

[img][/img]

If I change the position of the Rc & Cc, how can we explain the miller effect.

Does it have other effect on the whole circuit?

Thanks~

Title: Re: change the position of Rc&Cc for the miller effect
Post by rajkumar palwai on Nov 9^th, 2011, 8:58pm

Henry,
Miller effect doesn't change if u interchange the Rc & Cc positions. The circuit functionality remains same.

Title: Re: change the position of Rc&Cc for the miller effect
Post by henrytqy on Nov 9^th, 2011, 11:08pm

rajkumar palwai wrote on Nov 9^th, 2011, 8:58pm:

Henry,
Miller effect doesn't change if u interchange the Rc & Cc positions. The circuit functionality remains same.

Yes, I think so.
In integrated amplifiers if Rc is implemented with a mos transistor,in most cases it is usually place first to better control the resistance of the transistor. Do you know that why "better control"?

Thanks~

Title: Re: change the position of Rc&Cc for the miller effect
Post by rajkumar palwai on Nov 10^th, 2011, 1:19am

In the amplifer, we keep the Rc to cancel the zero caused by the gm2&Cc. So ideally Rc should track gm2 across process corners.

Now gm2 =K(Vgs2-Vth2). And we implement the Rc with a PMOS in linear region whose gds=k(Vgs-vth2). So if u connect the Rc and Cc in the order as shown in the figure, then the same node goes to the gate of M2 and source of Rc and hence u can control the Rc bias accurately.

Title: Re: change the position of Rc&Cc for the miller effect
Post by raja.cedt on Nov 10^th, 2011, 1:22am

hello,
There is no much diference...
may be first stage has lesser swing becaz of gm2, so if you use an mos as resister then Vgs may not change. and moreover control for the resister will come from replica bias, so to track better not to have any swing.

Thanks,
raj.

Title: Re: change the position of Rc&Cc for the miller effect
Post by RobG on Nov 10^th, 2011, 8:15am

henrytqy wrote on Nov 9^th, 2011, 11:08pm:

Yes, you think correct for a mos resistor. In that case, you want the mos resistor on the left. The resistance is controlled better there because the voltage at the "-" node of the gm2 stage varies only a small amount, thus the gate-source voltage of the mos resistor is relatively independent of the output voltage. On the other hand, the output of gm2 can swing rail-to-rail. If you want the mos resistor on the right side of the cap you need to make it a tgate. For example, if you just use an NMOS it will turn off when the output is low leaving you with no compensation.

Title: Re: change the position of Rc&Cc for the miller effect
Post by weber8722 on Nov 17^th, 2011, 7:59am

I would leave the positions as drawn :). If Cmiller is big, than also maybe its bottom-plate cap, so have this at the driven side - i.e. at the op-amp output!

Title: Re: change the position of Rc&Cc for the miller effect
Post by RobG on Nov 17^th, 2011, 11:15am

weber8722 wrote on Nov 17^th, 2011, 7:59am:

I would leave the positions as drawn :). If Cmiller is big, than also maybe its bottom-plate cap, so have this at the driven side - i.e. at the op-amp output!

Good point about the bottom-plate parasitic